RAI & Money God League: An Introduction to DPI-backed Stable Credit šŸ—æ

Background

Reflexer Labs has designed a new type of stable asset RAI, a self-stabilizing asset backed credit facility (more on self-stabilizing assets here).

After its launch in February, over $100m worth of RAI has been minted and more than $300M worth of ETH have been deposited in the protocol. RAI has managed to stabilize using its on-chain PID controller and is now undergoing a governance minimization process with the intention to harden the protocol.


More stats on RAI are available here

Reflexer Labs is now introducing the Money God League, an initiative meant to bring non-pegged stable assets into the mainstream and make control theory a common framework used in a wide variety of DeFi related projects.

The Money God League is specifically targeting strong, established communities that can benefit from running their own autonomous credit facilities. This post marks the first public announcement of The Money God League, a purposeful sign of support to the INDEX community and credit to its achievements.

The Index Coop as a platform and DAO is an ideal partner because of the effective community governance & the nature of Index products. These Index products would be a unique collateral source for a new RAI-like asset which could in turn unlock more Index product use cases and liquidity. This is why weā€™re extremely excited to announce the League!

:fire:_ :fire: (Fire Eyes DAO) is working alongside the Reflexer team to communicate and coordinate this partnership. :fire:_ :fire: is a DAO navigating token economic and governance systems across the DeFi ecosystem. As a collective, weā€™ve been supporters of both The Index Coop and RAI projects since inception and the ability to coordinate growth between them is something :fire:_ :fire: was built for.

Summary of this post

  1. Introduce RAI & the ā€˜DPI Creditā€™ structure
  2. Kickstart discussion between the Reflexer & The Index Coop communities
  3. Map out the longer term possibilities of collaboration

What is RAI (TL;DR)

:moyai: DeFi Lego

  • Fork of MakerDAOā€™s Multi-Collateral DAI (MCD)
  • Governance-Minimized in the Long Run
  • Algorithmic (PID) interest rates

:moyai: Self-stabilizing asset-backed credit facility

  • Like MakerDAO, users can unlock credit from their ETH
  • Unlike MakerDAO, debt/credit is not fixed at $1. Rather, it fluctuates based on supply and demand.
  • RAIā€™s PI controller updates the systemā€™s moving peg to balance demand for debt vs. credit
  • More capital efficient than money markets as capital cost is 0 (RAI is minted)

:moyai: Reserve asset for DeFi, alternative to ā€œstablecoinsā€

  • More decentralized than dollar-pegged stablecoins like USDC & USDT
  • Independent of USD inflationary monetary policy
  • Avoids being targeted by dollar coin regulation like the ā€œStable Actā€
  • Backed by pure ETH, no centralized collateral

Discussion

As a result of the launch of the Money God League, we want to steward the creation of a DPI-backed version of RAI. Weā€™d love to use this forum thread as a jumping off point for this project!

DPI-RAI Summary

  • DPI as collateral along with its community and governance structures lend themselves perfectly to introduce a ā€˜stable DPI-backed creditā€™. This would be a system similar to RAI where INDEX governs a stable credit backed by DPI (and potentially other index products in the future).
    • Introducing a ā€˜stable DPI-backed creditā€™ to the Index communities has three core benefits:
      • DPI community and wider DeFi users have a stable credit system to leverage, thus giving more utility to DPI and INDEX.
      • DPI gets a deeper liquidity pool, allowing for wider market reach and impact.
      • The INDEX community has full discretion and governance over adding new index products (and potentially DeFi tokens) as collateral and unlocking liquidity from them.

Discussion Points for the INDEX community

  • Would you like to see a volatility-minimized asset backed by DPI?*
  • What use-cases do you see for a DPI-backed stable credit? Lending, leverage, etc.*
  • What other indices assets apart from DPI would you like to see as collateral types for this new stable credit system?*
  • We would need a new name for DPI-RAI. What name could a DPI-backed credit have?*

Implementation

Initial setup and implementation

If the collaboration were to move forward, the following steps would be needed to deploy and support DPI-RAI:

  • A proposal would be made to INDEX governance, explaining the system, its requirements, benefits, risks and path to adoption.
  • In the event of a passing proposal:
  • The Reflexer team would deploy the new system, recommend controller parameters, carry out ongoing maintenance and share the necessary smart contract infrastructure with IndexCoop.
  • INDEX would govern the protocol, build and operate the UI, manage communications, and once a stable implementation is live on mainnet, take over the keeper roles.
  • Further discussion and proposals (to both FLX & INDEX communities) would be created around:
    • Liquidity incentive collaboration around the adoption of DPI-RAI.
    • Further social and economic alignment between the FLX and INDEX communities.

Long-Term Roadmap

  • Working to develop a Collateral Roadmap and a timeline to add new collaterals in DPI-RAI.
    • MVI, INDEX itself?, etc.
  • Creating an Incentives Roadmap for the new stable credit deployment, possibly introducing two sided incentives for the liquidity of DPI & DPI-backed credit.

Incentive Alignment

  • In the spirit of using DeFi to align incentives, we propose to use a wrapped 50/50 FLX/INDEX LP share as the governance token / fee burn token / insurance mint backstop for the system.
  • In MakerDAO the MKR token plays 2 roles: governance and the ā€œlender of last resortā€. If the system is undercapitalized, MKR is minted and sold until the system is recapitalized (as it did on March 12th).
  • As an incentive for MKR holders taking on the risk of backstopping the system, the system earnings (stability fees) from all collateral types that accumulate beyond some threshold are used to buyback and burn the MKR tokens via public auction.
  • In Reflexer, the FLX token plays the same role as MKR, and in the proposed jointly operated system we propose the wrapped FLX/INDEX LP share could play the same role.
    • The reason we would ā€œwrapā€ the the 50/50 FLX/Index LP Share is to make it work with minting / burning; the system would mint & burn the wrapper token, meaning that holders of the wrapped token would get less of the underlying LP shares if wrapped tokens are minted (lose value), and when the wrapped tokens are burned the holders of the wrapped token get more of the underlying LP shares (gain value).
  • Aside from governance and the lender of last resort use-case, the wrapped LP Share can also act as a ā€œlender of first resortā€. This means that tokens can be staked in a pool where they get sold off in case the DPI-RAI system is underwater and in exchange for this, the pool receives a portion of the surplus accrued by the protocol.
  • Weā€™re open to suggestions for better incentive alignment setups, this was just some ideas to kickstart the conversation!

Next steps:

Discussion! Thereā€™s a significant amount to digest above and weā€™re keen to get as much community input as possible. Questions are in italics, reply to individual sections, give us your thoughts!

  • The INDEX community providing discussion & feedback on the above ideas
    • Decide on collateral roadmap
    • Decide on governance roadmap
    • Propose LP incentives budget
  • Drafting an initial proposal

RAI Learning

:moyai: Website: https://reflexer.finance/
:moyai: Github: GitHub - reflexer-labs/geb: Core smart contracts for GEB
:moyai: Discord: https://discord.gg/3FEJX3h
:moyai: Twitter: https://twitter.com/reflexerfinance
:moyai: Docs: https://docs.reflexer.finance/
:moyai: Bankless - Trust Minimized Money (RAI Explainer)
:moyai: Stefan - RAI is Live (RAI Launch Post)
:moyai: Ameen - A Money God RAIses (RAI Launch Post)
:moyai: Defiant - Ameen on RAI (Podcast)
:moyai: ETHGlobal - Building a Money God (Presentation Recording)

12 Likes

Why wait for DPI on Maker when we can make our own maker?

Iā€™m super pumped for this, Iā€™m a big fan of RAI/Reflexer.

Some quick questions:

  • RAI is supposed to be governance minimized but do you plan for this to be a fully governed system?
  • If yes^ Do you have any suggestions for ungovernance? If no^ How do you plan on adding collateral types for different index products?
  • Would there be a stable asset for each index product or a single asset backed by all index products?

I think the main question for Index Coop are

  1. Does this distract us from our core offerings
  2. How can we leverage this in current and future products like using DPI-RAI in SDI instead of DAI or earning yield on SYI.

This could even be used for treasury management for a stable asset that still adds value to INDEX while reducing price volatility.

6 Likes

Weā€™re also fans of Index and want to see it grow exponentially :smile:

For your questions:

  • It depends on what the Index community wants. We already have the infrastructure to automate most day-to-day processes although I imagine that the community might want to retain governance rights over adding more collateral types (e.g other indexes). I think Index has a unique opportunity to build a stable asset backed by multiple indexes and possibly by DeFi tokens so I encourage everyone to think about it
  • Yes we do although it still depends on what the community envisions. We have all the tools, you decide what to do with them. We can collaborate with the Index community on thinking about params for new collaterals and showing you how to add them
  • I donā€™t recommend that level of fragmentation. Index could build a strong brand with one stable asset backed by multiple collaterals

I think the main advantages of having your own on-chain credit facility are:

  • Full discretion over setting stability fees and the way your PI/D controller works, thus attracting users/capital better than some other generic credit facility
  • You concentrate liquidity for all your products into one thing while your community still has exposure to all the collaterals they use to mint DPI-RAI
5 Likes

The incentives are well aligned for this to pass.
As for the name I propose
iRai or possibly
dRai

3 Likes

Hey @oxlucas and team - this is an exciting proposal. Thank you for putting it together and posting it to our forums! I believe this is a massive win for Index Coop. Furthering the use cases of $DPI and other products as collateral is a huge priority right now.

If you were not bringing us this proposal today - I firmly believe that we would be coming to you in the next few months asking to do the exact same thing.

This proposal does a few things for us.

  • It allows us to build highly customized leveraged Indices around the products that we own. By controlling lending and borrowing the levers we will be able to create vastly more leveraged indices and other exchange traded products.

*It allows us to quickly list new indices as collateral. This helps us bypass the often extremely lengthily collateral on-boarding process on major lending platforms.

  • Diversify our revenue streams and align incentives between two major and cutting edge DeFi protocols. This is true unforkable value (@Mringz )

@Kiba raises a valid concern here. Our community will need lots of help from your community initially to get this product up and running. We will likely need an ambassador from your community to help us run it for a while.

With that said - we are more than capable of executing this proposal and doing it well. This is the time for these exact kinds of ambitious DeFi partnerships!

3 Likes

thanks @0xLucas, @sionescu and the team for the proposal.

To sum it up, this proposal would allow the Coop to create a volatility-minimised stable asset backed by DPI and potentially other index products in the future. But we wouldnā€™t just be creating a stable asset but the ecosystem supporting it as well ie weā€™d be replicating Maker (with modifications pioneered by Reflexer with RAI) and using FLX/INDEX LP token in a similar way to MKR token in the MakerDAO ecosystem. Did I get that right?

On this point, any idea on how much we can leverage the existing reflexer.finance UI for this?

Would be great to understand governance requirements for the ecosystem. What are some of the governance parameters the community would have to decide on? Will we be deciding on things like stability fees, collateral ratio, liquidation penalties, etc.?

Overall, I love the idea and the concept of Index Coop running our own credit ecosystem (with help from reflexer and fire eyes DAO) is, frankly, mindblowing. Look forward to see how the conversation develops. @0xLucas as I said in the Discord chat, I think maybe a Discord AMA would be great to help our community wrap our heads around this.

3 Likes

Love it! Iā€™ve been following RAI since launch, and Iā€™d be excited to participate in this move as well!

Looks like I may have upcoming work to do :slight_smile:

For your first point, yeah thatā€™s right :mechanical_arm:

I fully agree that an AMA would be really useful so everyone is brought up to speed.

Technically we can share the full UI, practically I think you might want to add custom branding so Iā€™m not sure how useful it would be to take the RAI dashboard as it is.

The following is what we have in mind for governance.

Short term (1-2 months post launch)

In the short term (1-2 months post launch), we recommend either of the following:

  • The Index and Reflexer teams manage the whole DPI-RAI protocol using a multisig
  • The teams only manage the DPI-RAI governance token and any other contract that interacts or references the gov token + the contract that collects stability fees

This ensures that our teams can act quickly if thereā€™s anything wrong with the protocol. After the initial period, we can pass control to governance token holders.

Long term

After the initial multisig gov period, we propose the following:

  • Governance minimize a portion of the protocol to ensure that the wrapped INDEX/FLX LP token remains at the core of the system and no one can vote it out
  • Pass all the remaining governance to the community

The first point ensures that our communities remain aligned in the long run and no one can vote out the initial governance token. To this extent, we already made a set of contracts that will help and we can share and explain them in an official proposal.

In terms of active governance, the following are params/contracts that may need to change more frequently:

  • Stability fees (usually you can just set this once per collateral and leave it like that unless the risk profile of that collateral changes)
  • Contracts/adapters to allow people to add new collateral types in the system (we already have several adapters and can show the Index community how to add new ones)
  • Stability fee treasury parameters. The treasury is in charge with paying others to update oracles, update the PI controller etc. Youā€™d need to update reward amounts from time to time but weā€™re working on contracts that also automate this to a certain degree

There are a bunch of other params that need to be frequently changed (liquidation penalties, collateral debt ceilings, debt floors etc) but many of these changes can also be automated using contracts that weā€™re currently building. This means that youā€™d need to pay more for bots though (depending on update frequency).

2 Likes

Hey all,

Thanks for the awesome feedback so far! The response has been impressive and a testament to the Index community :tada:

This posts aims to identify key next steps for the Index & Reflexer community:

Community Discord AMA :speech_balloon:

  • Weā€™d love to get a community AMA setup where the IndexCoop community can have an open line of communication & question to The Reflexer team. Weā€™d be happy to coordinate this on our end as itā€™d be great to get Stefan, Ameen and :fire:_ā€‹:fire: into either a Discord channel or Zoom call mid next week?
  • In terms of timing, weā€™re mostly on US timezones. Anytime around lunchtime PST seems like the most conducive time, encompassing all USA, evening for europe, early morning for Asia/Australasia.
  • Proposed times:
    • 12pm PST Wednesday 21st April
    • 12pm PST Thursday 22nd April
    • 1pm PST Thursday 22nd April

Index :handshake: Reflexer Core Contributor Call

  • Itā€™d also be great to coordinate a call with a handful of the Index core contributors and developers with the Reflexer team. Weā€™d love to run through some of the more technical/nuanced details and begin mapping out a game plan for this integration, working towards achieving the following outcomes:
    • Formalizing a comprehensive list of requirements/dependencies involved in setting up and maintaining the system.
    • Assigning ownership of certain areas/tasks to each community and opening proper comms channels.
    • Getting a formal proposal ready to submit to INDEX governance in 3-4 weeks.
    • Buidling and beyond!

Naming the stable-credit :white_check_mark:

  • Also just wanted to collate the current list of possible DPI-backed stable-credit names: DPAI, DEFAI, DPRAI, dRAI, iRAI, dRai
    • Please keep them coming!

Let us know if you guys have any further questions and concernsā€“weā€™ll be monitoring this thread regularly. Super stoked on this proposal and itā€™s responses already!

Letā€™s build stable credit for DPI :rocket:

9 Likes

I totally get the benefits of being able to use DPI as collateral for borrowing, and I will certainly use it as such. I am also a big believer of decentralized stable coins systems and following the RAI project with interest.

However what is not clear for me is what would be the benefits for Index Coop of running its own credit system?

I mean, it seems to me that using DPI on Maker to generate DAI, or DPI on Reflexer to generate RAI (I donā€™t know if there are any plans for this) would bring the same benefits to the DPI holders, with a lot less investment in time, resources, risks, etc, for the Index Coop.

I think the proposal should highlight what is the value added of ā€œIndex Coop deploying efforts to run its own credit systemā€ vs ā€œIndex Coop facilitating the use of DPI in existing credit systemsā€.

We have to ensure that any effort diverting from the core business of Index is worth it.

2 Likes

Happy to help coordinate here! As a first step, I am going to post those times in our Discord so folks can vote on what is best for them.

Sharing for context, here is how weā€™ve done community calls in the past: Synthetix, Bankless. That is not to say you canā€™t run it yourself or do it differently :slight_smile:

@James - what is your Discord handle?

I can help out here too

Very good question!

I wrote a bit about advantages at the bottom of this:

The intent behind DPI-RAI is not to limit it to DPI but rather be a credit facility for multiple indexes (and maybe even DeFi tokens).

The Index community controls the parameters set for each collateral and can also add new collateral types faster than any external credit facility.

If the plan was to allow only DPI holders to lever then I agree it would be easier to just add it in a separate credit facility. This proposal outlines a way bigger vision though where you have your own credit pipeline that INDEX can use to solidify its position in the market.

1 Like

Adding to @sionescu. Instead of having one or a few ā€˜central banksā€™ like Maker, this strategy makes index coop its own ā€˜central bankā€™ in a way. Although, it is tied to FLX so not entirely its own ā€˜central bankā€™ at the same time.

This sounds exciting and had some questions here about it. So from what I understand about the incentives - a wrapped token of the 50/50 FLX/Index LP Share is equivalent to the MKR token in the Maker system.

(1) Am I to assume thereā€™s going to be a control system implemented for the 50/50 FLX/Index LP wrapped token to ensure a more dynamic supply policy with some kind of invariant or conservation/reserve ratio?
(2) I assume similar wrapped tokens will be created with other notable DeFi projects that will have the ability to mint and burn. The concern here is the single point of failure that is the FLX token. Is there a risk mitigation plan for this? For example, if you have FLX pair with gov tokens from notable DeFi platforms, what risks are associated with FLX such that if there are any it could possibly affect these wrapped pairs causing a spill over effect in their credit facility systems? (Not sure if I communicated this properly but please let me know Iā€™ll try re-wording it).

Thanks in advance.
witwiki

Perfect! Iā€™ve sent you a message on discord @anon10525910 :pray:

The above context on community calls looks like a great structure, is it worth either of us posting to the forum with a solidified community call time/structure similar to the Synthetix & Bankless posts?

For everyone else, encourage you to vote on the times in the #annoucements channel on discord!

Iā€™m also conscious of the timeline here; Wanting to ensure we have enough time for the INDEX community to decide on a time & make plans to attend!

Weā€™ll also chat on discord about the core contributor call @anon10525910

1 Like

For 1), thereā€™s a simple smart contract that anyone can use to deposit LP tokens and get the wrapped version or burn the wrapped tokens to get LP tokens back. No need for anything else because the ratios between LP tokens and wrapped tokens are computed by the contract depending on the supply of each token.

For 2), the beauty is that each of these systems is isolated. If one of them doesnā€™t work out, neither INDEX (replace with any other protocol/governance token) nor FLX get inflated. The wrapped versions of the LP tokens do get printed, meaning that less LP tokens can be redeemed per one wrapped token.

TL;DR the LP + wrapped functionality allows for experimentation with different collaterals/controller settings without compromising the underlying tokens.

1 Like

ok,

Iā€™m not a financial expert, and Iā€™m probably sitting towards the middle of the bell curve, so I have lots of questions:

  • After RAI why is there a need for a second stable product?
  • Why is DPI considered good collateral for this (and not say LINK with 100x the market cap)?
  • DeFi is permission-less, there is nothing stopping the Reflexer team from launching DPI-RAI with no input from INDEXcoop.
  • What do reflexer labs gain from a joint implementation with INDEXcoop?
  • What do Indexcoop gain?
  • If the goal is governance minimisation, how is merging the activities of 2 DAOā€™s the most logical starting point?
  • How is DPI-RAI better than working to add DPI as collateral on Maker / AAVE / Fuse? - We get a liquidity pool that we can control for leverage, but at the cost of more complexibility.
  • I see you have proposed something similar to BargerDAO for a wBTC based RAI. How many different RAIā€™s do you forsee?
  • I thought that the roadmap was 100% ETH backed RAI? If so, why the change to multiple RAIs?
  • RAI currently has ~$100m issued and an income of 2% (i.e. $2 m ~ 0.2% of FLX FDV) for FLX holders. Correct?
  • The $100 M RAI issued is supported by the recent $40 m rewards program for the first 60 days of project being live?
  • What is RAI used for? I see that 90% of it is sitting in Uniswap earning about 6% trading fees.

Apologies for all the Mid IQ questions.

3 Likes

Happy to elaborate on all of these!

  • The type of collateral you use in a system like RAI will determine (to a certain extent) the behaviour of that stable asset. Dampened ETH != dampened basket of indices, hence the exposure you get when using RAI is different than the one from DPI-RAI
  • Due to the exposure it gives to DeFi tokens (although nothing stops Index from adding other individual tokens in DPI-RAI). As an aside: I think the name DPI-RAI might be a bit misleading. The idea is to start with DPI but then allow the Index community to add more indices in this system. Right now I can hold DPI or some other index that youā€™re issuing but it would be even more useful to hold exposure to an index, use it to mint DPI-RAI and then use that asset in DeFi
  • It would be counter-productive to our mission. We think Index has way more visibility and expertise in analysing DeFi tokens and building indices so it makes sense that you run a credit facility backed by these collateral types. We want to see more RAI like assets in the wild and we think that helping other communities with their own specialized credit facilities is the best way to achieve this goal
  • Wrote a bit about it in this answer
  • The goal for RAI specifically is governance minimization. If the Index community thinks DPI-RAI should also be governance minimized at some point thatā€™s great, we can help with that! The idea is that we come together and bring different skill sets that help us achieve the vision of an index backed credit facility
  • Wrote a bit about this here. If the Index community only wants to use DPI and not have any other index in a credit facility, then I agree you can simply add DPI in another protocol. This proposal is for a RAI like asset backed by multiple indices and the community can fine-tune parameters for each of them
  • ETH, BTC, indices/DeFi tokens. These are the major ones I see and imo each could have its own credit facility
  • RAI will indeed be backed solely by ETH although from the beginning people have been asking us about other collateral types and we mentioned that creating other systems like RAI is something weā€™d like to pursue
  • Yup
  • TVL got to about $390M and our RAI/ETH pool had more than $200M at some point. We scaled down incentives in order to focus on other areas (e.g RAI adoption on secondary markets). Youā€™re right that about $40M was distributed in the first batch
  • RAI must have plenty of liquidity on exchanges in order to be hard to manipulate. Weā€™re working on a Uniswap v3 liquidity manager so we can focus capital around the moving peg as well as scale down the LP program. RAI is also on CREAM, Fuse and from next week on RULER and it will soon be on L2 and even more protocols. BD for a stable asset takes time :slightly_smiling_face:

Hope this was helpful, let me know what points I should expand!

3 Likes